Spaces:
Sleeping
Sleeping
| import streamlit as st | |
| from transformers import pipeline | |
| # Load the Hugging Face models | |
| classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli") | |
| sentiment_analyzer = pipeline("sentiment-analysis", model="distilbert-base-uncased-finetuned-sst-2-english") | |
| # Define the categories for customer feedback | |
| CATEGORIES = ["Pricing", "Feature", "Customer Service", "Delivery", "Quality"] | |
| # Function to map sentiment to a rating (1 to 5) | |
| def sentiment_to_rating(sentiment_label, sentiment_score): | |
| """ | |
| Convert sentiment analysis results (label and score) to a rating from 1 (most negative) to 5 (most positive). | |
| """ | |
| if sentiment_label == "POSITIVE": | |
| if sentiment_score >= 0.8: | |
| return 5 # Most positive | |
| elif sentiment_score >= 0.6: | |
| return 4 | |
| elif sentiment_score >= 0.4: | |
| return 3 | |
| else: | |
| return 2 | |
| elif sentiment_label == "NEGATIVE": | |
| if sentiment_score >= 0.8: | |
| return 1 # Most negative | |
| elif sentiment_score >= 0.6: | |
| return 2 | |
| elif sentiment_score >= 0.4: | |
| return 3 | |
| else: | |
| return 4 | |
| # Streamlit app UI | |
| st.title("Customer Feedback Categorization and Sentiment Analysis") | |
| st.markdown( | |
| """ | |
| This app can detect the topics and intent of customer feedback | |
| and determine the sentiment (rating from 1 to 5) for each relevant category. | |
| """ | |
| ) | |
| # Input text box for customer feedback | |
| feedback_input = st.text_area( | |
| "Enter customer feedback:", | |
| placeholder="Type your feedback here...", | |
| height=200 | |
| ) | |
| # Confidence threshold for displaying categories | |
| threshold = st.slider( | |
| "Confidence Threshold", | |
| min_value=0.0, | |
| max_value=1.0, | |
| value=0.2, | |
| step=0.05, | |
| help="Categories with scores above this threshold will be displayed." | |
| ) | |
| # Classify button | |
| if st.button("Analyze Feedback"): | |
| if not feedback_input.strip(): | |
| st.error("Please provide valid feedback text.") | |
| else: | |
| # Perform zero-shot classification | |
| classification_result = classifier(feedback_input, CATEGORIES) | |
| # Filter categories with scores above the threshold | |
| relevant_categories = { | |
| label: score | |
| for label, score in zip(classification_result["labels"], classification_result["scores"]) | |
| if score >= threshold | |
| } | |
| if relevant_categories: | |
| st.subheader("Categorized Feedback and Sentiment Ratings") | |
| # Perform sentiment analysis for the feedback | |
| sentiment_result = sentiment_analyzer(feedback_input) | |
| sentiment_score = sentiment_result[0]["score"] | |
| sentiment_label = sentiment_result[0]["label"] | |
| # Display results for each category | |
| for category, score in relevant_categories.items(): | |
| sentiment_rating = sentiment_to_rating(sentiment_label, sentiment_score) | |
| st.write( | |
| f"**Category**: {category}\n" | |
| f"**Category Score**: {score:.4f}\n" | |
| f"**Sentiment**: {sentiment_label} ({sentiment_score:.4f})\n" | |
| f"**Rating**: {sentiment_rating}/5" | |
| ) | |
| st.markdown("---") | |
| else: | |
| st.warning("No categories matched the selected confidence threshold.") |